Abstract: Systems and methods are provided for predicting the effects of acceleration on a probability of a successful completion of a project. Specifically, one or more machine learning algorithms can be trained to predict the success of a project from a plurality of features, including at least one feature that is a function of an end date of the project. A set of projects can be selected from projects having expected end dates that do not fall within a desired window of time, each project having a first probability of success given the current end date. A second probability is calculated for each project at a machine learning platform that quantifies a chance that the project will be successful if completed within the window of time. A difference between the first probability and the second probability is determined for each project, and the projects are selected according to the difference.